Solar water splitting has shown promise as a source of environmentally friendly hydrogen fuel. Understanding the interactions between semiconductor surfaces and water is essential to improve conversion efficiencies of water splitting systems. TiO2 has been widely adopted as a reference material and rutile surfaces have been studied experimentally and theoretically. Scanning Tunneling Microscopy (STM) is commonly used to study surfaces, as it probes the atomic and electronic structure of the surface layer. A systematic and transferable method to simulate constant current STM images using local atomic basis set methods is reported. This consists of adding more diffuse p and d functions to the basis sets of surface O and Ti atoms, in order to describe the long range tails of the conduction and valence bands (and, thus, the vacuum above the surface). The rutile TiO2 (110) surface is considered as a case study.